A Survey Paper on Software Bug Classification Techniques using Data Mining

Abstract

A Software bug is a blunder, blemish, disappointment or deficiency in a PC project or framework that causes it to deliver an off base or surprising result. At the point when bugs emerge, we need to settle them which is difficult. The greater part of the organizations burn through 40% of expense to settling bugs. The procedure of altering bug will be bug triage or bug collection. Triaging this approaching report physically is blunder inclined and tedious .programming organization pays the greater part of their expense in managing these bugs. In this paper we arranging the bugs with the goal that we can decide the class of the bug at which class that bug is has a place and in the wake of applying the order we can dole out the specific bug to the precise designer for altering them. This is effective. In this paper we are utilizing mix of two grouping strategies, guileless bayes (NB) and k closest neighbor (KNN).In advanced days organization utilizes programmed bug triaging framework yet in Traditional manual Triaging framework is utilized which not effective and setting aside an excess of time .For is triaging the bug we require bug subtle element which is called bug store. In this paper we likewise diminishing the bug dataset in light of the fact that on the off chance that we having more information with unused data which causes issue to relegating bugs. For actualizing this we utilize occasion determination and highlight choice for lessening bug information. This paper portray the entire methodology of bug assignment from beginning to end and finally result will appear on the premise of chart .Graph speaks to the most extreme plausibility of class means at which class the bug will has a place.